Content-Type: text/html

Introduction

The title of this paper, couched as a 'report from the field,' is somewhat
consciously and playfully ironic. The metaphor of 'field' suggests a
going-out-into-the-world in search of empirical facts and evidence about some
phenomenon, much as would the ethnographer, archeologist or anthropologist, and
that upon returning home one is prepared to report one's findings or
impressions. But for the textual critic, discourse itself is the field - a
textual field - into which the analyst ventures in search semiotic clues, signs,
symbols, linguistic structures or other phenomenon that collectively constitute
the meaning of the text. Construing discourse analysis as a kind of field
research, identified generally with qualitative content analysis, draws on the
same metaphor that Glassner & Corzine (1982) use in envisioning library research
as "fieldwork." The field metaphor is also used widely with respect to the study
of 'intellectual' fields or the competing professional knowledge domains in
which social and political argumentation occur - "in which power and capital are
deployed" (Prosise et al., 1996). Social field theory draws heavily on the work
of Bourdieu (see, for example, 1977, 1984, 1985, 1990, 1991), who focuses on
"social agents" as they are constitutive of  "symbolic forms of social authority
and the discursive struggles at the root of the structured forms of social
authority" (Prosise et al., 1996, p. 119).
     In the present paper, both of these related notions of field are useful as
a general intellectual framework for the comparative analysis of two newspaper
texts. The present paper is a report from the field because it is extracted from
a longer work in exploratory news-discourse analysis (Harry, 1999), in which a
much larger number of comparative newspaper texts are examined with respect to a
long list of discursive features. As a way to provide a demonstration of
discourse analysis via textual exemplars, this paper compares only two articles
from the larger sample, and attempts to map several (but by no means all)
important discursive features that collectively constitute the discourses of
each text. In other words, I am more interested here in an in-depth analysis of
standard discursive features that frequently can be found in many news texts,
rather than an analysis that, taking in a larger number of articles, would
provide more breadth. Because the two articles are essentially framed from the
perspective of the news sources (social agents) quoted, the discourse analysis
also encounters the respective intellectual fields or knowledge domains of those
competing sources whose views are presented so prominently in each article.
     Discourse analysis is a form of linguistics that differs essentially from
content analysis not only in the former's qualitative (rather than quantitative)
focus, but also in its attention to the overarching grammar of a text, how each
sentence relates systematically to present one or more central themes or topics
that can be interpreted as textual meaning. Discourse analysis is intensively
language-focused. As a methodology, it attempts to understand, holistically, an
entire text or entire texts through systematic analysis of the texts specific
language, rather than to 'read off' either the meaning or the central
characteristics of a text by examining only how often certain words, phrases or
linguistic patterns occur, as in much quantitative content analysis.

As Jensen (1987, p.9) notes:
The decisive advantage of discourse analysis_is that it allows the researcher to
study variations in language use that are relative to the social context of the
communicative interaction in a systematic way. As a result, it is comprehensive
enough to include analyses of, for example, an interview between a politician
and a journalist, the inferences made on the basis of this interview in a
relatively compressed news item, and the conclusions the reader/viewer draws
about the politician on the basis of this interview. In each case actual
linguistic patterns, including lexical choices, grammatical constructions, and
story coherence, have an explanatory value and serve as the point of departure
for describing the more fundamental social mechanisms at work in each stage of
the communication process.

Outline of Paper
In this paper, I first attempt to ground my own theoretical perspective within
previous discourse-analysis techniques deployed by others studying both print
and broadcast news stories. I then briefly describe the historical background of
the two news articles analyzed, as a way of providing relevant context for
understanding the broader environmental conflict that both newspapers were
reporting. I next provide a detailed discourse analysis of  relevant portions of
each comparative news article portraying the same event - a protest by those
opposed to the construction of a local hazardous waste-treatment incinerator
during a federal permit hearing - and provide several figures that demonstrate
how these texts can be discursively 'mapped.' Finally, I offer some general
speculations and conclusions about the nature of these articles and what they
may say about the functioning of ideology in news discourse.
     Ultimately, what I attempt to provide through this technique is an
admittedly exploratory system (my own) of what might be thought of as
'ideological mapping' - an analysis of how any news article reproduces the power
of news sources and, consequently, asserts and sustains the institutional power
of journalism itself as a sense-making institution devoted to constructing a
'discourse of the real' (Jensen, 1986). The discourse mapping provided in
Figures 1, 1a, and 2 draws, in part, on the schema categories - the syntactic,
narrative conduits through which semantic information is deployed within a news
texts - as developed by van Dijk[1]
Literature Review
 Although it has been available as a well-documented technique since at least
the 1970s, discourse-analytic work has never gained a foothold in analyzing
newspaper or broadcast journalism in the American context. Fairly popular among
British and Scandinavian researchers (see especially Fowler, 1991, and Fowler,
et al., 1979; Hodge, 1979; Trew, 1979), news discourse analysis is all but
absent as a methodology for examining American journalistic language, the
favored technique being standard, quantitative content analysis; a literature
search for examples of news discourse analysis applied to American news media in
the last 10 years produced not a single scholarly example. Discourse analysis,
however, is a term so widely used, so often loosely applied in literary,
cultural, critical, sociolinguistic and functional-systemic interpretive studies
(for representative examples of the latter two research domains, see especially,
Halliday, 1985; Martin, 1992; Ventola, 1987; and Steele & Threadgold, 1987), and
so widely targeted to all manner of written, visual and oral texts, that the
term itself risks losing any theoretical specificity. Practically speaking, it
has come to mean that one somehow analyzes discourse - period. In this sense,
discourse analysis can hardly be distinguished from content, thematic, narrative
or various species of literary interpretation, none of which have anything,
necessarily, in common. What usually distinguishes discourse analysis as applied
to news, and which I will follow in the present paper, is its focus on specific
textual patterns based in both wording and sentencing structures, as understood
within a larger social-political framework or social structure. In this respect,
the methodology of discourse analysis also implies a theoretical stance,
grounded in the notion that discourse reflects social structure and thus,
reflects and refracts predominant ideologies in and through the text. As van
Dijk (1983, p. 27) states, "_a discourse analysis will be part of a more
embracing cognitive and social theory about the rules and strategies that
underlie the production and understanding of media discourse_[S]uch work
attempts to uncover implied meanings that represent ideological positions"
(1983, p. 27), and it demonstrates how news routines structure source views,
and thus present "possible 'formulations' of reality" (1983, p. 28).
     In the present study, the discourse analysis technique applied to the two
comparative newspaper stories - regarding a local environmental conflict
reported by the Pittsburgh (Pa.) Post-Gazette and the East Liverpool (Ohio)
Evening Review - draws on the general principles of news discourse analysis
recommended by van Dijk (1983, 1985, 1988a&b, 1991a&b, 1993), by Jensen (1986,
1987), and by Fowler (1991), and Fowler et al. (1979). Each theorist approaches
news discourse analysis in somewhat unique ways based on respective theoretical
interests. For example, van Dijk tends to be more interested in linguistic
structures, Jensen in questions of the political economy of news, and Fowler and
company with questions of power and ideology, though each scholar is generally
interested in all of these domains. Concentrating primarily on press discourse
from European newspapers, Van Dijk (1983, p. 21) analyzes "structures and
functions of actual forms of language use, that is_discourse," and ties his
study of  linguistic propositions, basic themes and syntactical schemas to
similar interpretive, psychological schemas or interpretive frames deployed by
news audiences (1983; 1988a&b). Similar to Jensen, van Dijk contends that
"surface structure" of language "not only expresses or indicates social
structure, but also, and even primarily, is meant to express underlying meaning"
(van Dijk, 1983, p. 25). This notion gets at something like the existence of a
textual hermeneutic, a meaning that can be prized from beneath the surface of
the sentences when these sentences are read collectively, as forming the
discursive essence of the text. In literary and film studies, Jameson's (1982)
notion of the 'political unconscious' of the text is similar, although in
Jameson narrative analysis tends to remain much more abstract, philosophical and
historicist, given its post-Marxist, deconstructivist aims. The importance of
van Dijk's emphasis on extracting the meaning of the text is that, given a set
of systematic procedures that map various textual structures, one is in a
position to comment upon not only these discrete structures but, finally, the
meaning of the entire text. The emergent interpretation gains its validity from
the ability to refer back to specific features of this discourse mapping, and
will therefore make for more than just an off-the-cuff set of opinions about
what the text actually means.
     Jensen (1986, 1987) is somewhat more concerned with news language and the
political-economy of news and other social institutions. Directing his attention
to network television news discourse, he attempts to show that "the working
assumptions which guide the activity of news organizations set firm limits on
what is, or even could be, reported in the news," and concludes that a "new
version of the same social model is constructed every day, contributing to the
maintenance and perpetuation of a political and economic order" (Jensen, 1987,
p. 8). Jensen relates his own work to Tuchman (1978), who speaks of a "web of
facticity," a professional, interpretive news framework restricted to analyzing
socially authorized source views based on professional, objectively verifiable
statements, if only to avoid lawsuits. This latter avoidance technique is the
basis for Tuchman's now famous formulation of journalism as a "ritual of
objectivity," a set of norm-based professional reporting practices that assure
the stability of the journalistic institution, and which result in the news
media ultimately reproducing status-quo interests [see also, Tuchman, 1977].
Along this line, Paletz et al. (1971) found evidence in analyzing comparative
local and national news media coverage that the news media support "local
government authority." Turow (1985) reaches much the same conclusion about the
status-quo mentality of news media, but includes the important qualification
that the news media, to fulfill their social-responsibility mission of providing
contrasting views - as well as to provide colorful, dramatic, conflict-rich
stories of interest to a mass audience - engage in 'cultural argumentation.' In
this sense, the news media both support and critique social power arrangements
and social agents. This dual function of the news media is acknowledged, as
well, by Gitlin (1980), who nevertheless concludes that the social critique via
news media stories is a smaller part of the larger game of reproducing the
dominant ideology.
     Studying mainly British newspapers and working within a broadly Marxist
'critical discourse' perspective, Fowler (1991) and his colleagues (Fowler et
al., 1979) dig deeply into newspaper language by studying frequently occurring
words and synonyms as well as discrete sentences said to reflect the larger
ideological interests of a given newspaper. For example, Fowler (1991) studies
whether a headline or lead sentence is written in the active or passive voice.
Determining active or passive voice in a sentence is thought to indicate how
directly responsible various social agents are for a variety of acts reported
about, and this is seen as a way to track the otherwise latent ideology at work
in newspaper stories.
     But what each of these theorists share (especially van Dijk, Jensen and
Fowler and his colleagues, all of whom work specifically within the
discourse-analysis tradition) is an interest in examining closely the actual
language of the news as reflective of not only the ideological or
political-economic interests of news sources, but also as to how these interests
are reproduced by news reporters themselves, who represent and are thus
constrained in what they can report by the professional norms and practices of
journalism as part of the mass-media institution. In the broadest sense, the
analysis of news as discourse is the analysis of how various social agents
(including reporters) - representing different and often competing
political-economic interests - cohere in a news text or set of texts. The
resultant analysis will draw out the systematic connections in lexis (specific
words and phrases and their meanings), and in the syntactical structures (how
one sentence connects to others in some systematic fashion to provide a textual
grammar). From a critical discourse-analysis perspective, the analyst will be in
a position to identify how the ideological views of sources are constructed by
journalists, thus how journalism itself is an ideological practice.
     In this paper, I use the term 'ideological problematic,' borrowed from
Althusser (1971)   as a textual marker indicating where in the news text a
conflict or thematic problem is articulated. As I hope to show, the ideological
problematic exists as a mixture of source views and reporter description,
analysis and metaphor. The ideological problematic is an important addition to
news discourse analysis because it points to how an interpretation can be
galvanized to specific nodes or places in a given text, so that the essential
problem or conflict grounds the rest of the text.
Contextual Background
The two news articles examined in the present study are part of 74 comparative
same-day/same-event news stories (37 each in the Pittsburgh and the East
Liverpool, Ohio, newspapers) that comprise a much more lengthy study (Harry,
1999). These 74 articles are part of an even larger universe of nearly 400
individual news stories (N=389) written by both papers between 1991 and 1993.
The Pittsburgh paper, a seven-day daily with multi-county distribution and a
daily circulation of about 250,000, wrote 107 staff-penned articles. The East
Liverpool paper was, during those years, a six-day-a-week evening paper with
single-county distribution and daily circulation of about 11,000; it has for the
last approximate two years gone to morning circulation. The Pittsburgh paper has
a reporting staff of at least several dozen regular reporters, while the East
Liverpool paper maintains a reporting staff of no more than four or five
full-time reporters. From a resource point of view, the Post-Gazette is better
equipped to provide a wider range of news reporting on a greater diversity of
events, and each reporter has the luxury of not always producing a story every
day; the typical small-town reporter is a multi-task performer, typically
producing several stories each day just to fill the 'news hole,' and thus having
less time to devote to any single story.
     Finally, the social structures of each newspaper differ dramatically:
Pittsburgh is a major metropolis with a diverse, ethnically rich local
population of roughly half-a-million, thus a naturally conflictual, impersonal,
heterogeneous social environment; from a communicative perspective, it is a
'pluralistic' news environment (Tichenor et al., 1980) with many competing
voices and a sustained attention to conflict and, thus, to a critical stance on
events. East Liverpool, about 40 miles due West in Ohio, is a largely rural
environment with some suburban outposts, and with a local population of
approximately 13,000; from a social communication perspective, the newspaper is
more naturally consensus-driven, less willing to report overt social or
political conflict in the interest of sustaining the more interpersonal,
homogeneous social environment (Tichenor et al., 1980)[2] of which it is a part.
     The conflict portrayed in the two news articles featured in the present
paper centers on a highly vocal, rancorous demonstration that occurred September
27, 1991, during the early phase of construction of the Waste Technologies
Industries (WTI) hazardous waste-treatment incinerator. Local protests had
occurred as far back as 1981, when investors first began pooling their resources
and planning to line up federal and state regulatory permits for the plant. Due
to legal hang-ups and financing and ownership struggles, the permitting process
stretched into the late 1980s. By the time it was evident that the plant had
secured its initial construction permits from the state and federal
Environmental Protection agencies (EPAs), local protest actions became more
organized. By 1991, local and regional plant opponents had joined forces with
the international environmental protest group, Greenpeace, whose media
event-centered oppositional strategies, tactics and organizational expertise
soon resulted in much wider local, state, regional, national and even
international news media attention.
     Although the East Liverpool Evening Review had written about 60 stories
before 1991 (the first ones dating back to the early '80s), its coverage
intensified dramatically when the plant got under construction in 1991. The
Pittsburgh Post-Gazette, while having run occasional wire stories in previous
years, never produced its first staff-written WTI article until Hollywood actor
Martin Sheen led a Pittsburgh-to-East Liverpool protest "caravan" in early June
1991. Interviews with two Post-Gazette reporters confirmed that Sheen's
appearance on the scene was the catalyzing event in the newspaper's decision to
begin covering WTI. A return appearance by Sheen in another protest, this one in
mid-October 1991, resulted in Sheen leading several other incinerator opponents
over WTI's wire-mesh fence, a protest stunt dutifully photographed and reported
on in the next days' newspapers, not to mention by broadcast media. This October
protest led to Sheen and several cohorts' arrests for trespassing on WTI
property.
     This, then, is the historical context within which the following
comparative articles of Sept. 27, 1991, can be understood. The June protest
caravan, which Sheen led, had occurred three months earlier, and the Sept. 27
articles analyzed here both noted, and used as a news peg, the fact that a
follow-up protest and return appearance by Sheen was scheduled in just a few
weeks, in October. So the Friday, September 27 articles (documenting a hearing
that had happened two days earlier, on a Wednesday evening) occurred relatively
early in the active protest/demonstration phase of the ongoing environmental
conflict, and are sandwiched between the previous June protest and the upcoming
October demonstration.[3]
     Comparative Discourse Analysis of the EPA Permit Hearing
The following analysis concentrates on headlines and lead sentences, or the main
topic that serves, schematically, to summarize a news event; on how Context,
Background and Verbal Reaction, three other of van Dijk's (1985; 1988a&b) schema
categories, syntactically structure the semantic or lexical content (words and
phrasings) of the news reports; and on key words and phrases, linked by a series
of connecting arrows that provide for a textual mapping. Rather than reproduce
full sentences, Figures 1, 1a and 2 [see below] reduce many sentences to their
bare essentials, primarily subject/verb structures and accompanying contextual
material. This sentence-reduction process follows Van Dijk (1988a).
     The ideological problematic, discussed in an earlier section and appearing
inside dotted lines in Figures 1, 1a and 2, identifies the conflictual theme
introduced by the reporter's own analysis of the event, based in part on the
reporter's use of source-centered material but formulated, as well, on reporter
interpretation via description and metaphor. Another aspect, which occurs in the
East Liverpool newspaper text, is the use of reporter and source-centered
strategies. Also indicated in one of the texts (of the Post-Gazette) is the
occurrence of an argumentative claim that emerges from source-quoted material.
     At the lexical (content-specific) level, arrows are used to demonstrate key
words, phrases and their synonyms. This is a major part of the mapping of a news
text, as redundancy - word and image repetition - is a fundamental aspect of the
mass-mediated text, designed as it is for consumption by a heterogeneous mass
audience (Vivian, 1997). Redundancy also serves to limit word and phrasing
choices, most of which are linked to official news sources, but some of which, I
will show, are constructed as descriptive/metaphoric devices by the reporter;
this latter function is the reporter's own ideological analysis, though it is
couched for the most part within the officially sanctioned views of sources.
Headline & Lead-Sentence Structures
Both news stories are opposition-focused. In other words, it was the protest
action of WTI opponents, who "chanted down" an EPA hearing officer
(Post-Gazette, paragraph one) and ultimately "forced federal and state officials
to halt a hearing into revised and modified permits for the project" (Evening
Review, paragraph 5), that motivated the reporters' own framing of what happened
at the hearing. However, each paper deals with this protest action in
significantly different ways. An examination of the headline/lead-sentence
structure in both papers reveals significant differences in what the reader's
attention is directed towards.
     Figure 1 shows that the Evening Review focuses in its headline on the fact
that Martin Sheen is scheduled to return for another protest: "Martin Sheen due
at WTI protest," and the lead sentence merely repeats the information,
indicating Sheen will return "to support the fight against construction" of the
plant. Comparing the headline and accompanying lead to that of the Post-Gazette
(Figure 2), a dramatically different storyline is constructed. Here, in the
headline, the incinerator "foes" are said to be on the march, their aim being to
"derail" the project: "Incinerator foes in Ohio plan march; Group wants to
derail East Liverpool project." Yet the lead sentence, surprisingly, has nothing
at all to do with the information featured in the headline, an occurrence that
Soley (1992), Strentz (1989), Marquez (1980), and van Dijk (1988a&b) all contend
can indicate newspaper bias. Recall that the schematic/syntactic function of the
headline and lead (van Dijk, 1988a&b) is to summarize for readers the major
theme of the story, the headline encapsulating what the lead fleshes out. Where
the lead and headline are consistent, informationally, in the Evening Review,
the headline and lead have no informational relationship at all in the
Post-Gazette. It is only much later in the latter article that the reader gets
an idea of where the concept of "derailing" the plant might have come from: The
reporter notes - in what amounts to a topic shift [Topic 2] that an "expert" on
environmental social movements says that protest movements "are often successful
in derailing projects" (Post-Gazette, paragraph 6). But nowhere in the story is
any source quoted as saying that WTI opponents want to "derail" the project; in
fact, the word is used by the reporter to characterize what can happen when
social movements are successful, the "derailing" of  a project. Whether the
quoted social-movement expert actually used this or a synonym we can only guess,
but the reporter ties this information to the quoted "expert."
     It becomes clear that the second part of the Post-Gazette headline, about
the "group" wanting to "derail" the "project," is the result of a headline
writer borrowing the reporter's word, "derailing," and actually turning it into
an unattributed claim: group wants to derail project. Whether the "group"
actually has such plans is irrelevant; no such claim appears anywhere else in
the news story. Thus the summary information in the Post-Gazette offers the
first portion of the article's ideological kernel - the problematic or conflict
that thematically infuses the text with a contradiction; here, I use
contradiction in the sense of a dialectic, one thematic element contrasted
against another.
     The Post-Gazette's lead sentence features the "chanting down" of the EPA
"hearing officer" by "700 opponents," and indicates that such action, according
to the hearing officer, left him with "no choice but to adjourn the hearing":
"An environmental regulator who was chanted down by about 700 opponents of an
Ohio hazardous waste incinerator said yesterday that he had no choice but to
adjourn the hearing Wednesday night." By contrast, the Evening Review does not
even mention the fact that the hearing was shut down until the fifth paragraph
of its story. This burying of what, from a journalistic news-value perspective,
clearly would be the catalyzing event of the hearing (the hearing that never
happened), is all the more surprising because the hearing took place in East
Liverpool, the Evening Review's headquarters.
     The September 25th hearing was a Wednesday evening, but there is no record
of a September 26 (next-day) story reporting the hearing in the Evening Review.
The first record of any Review story is not until Friday, September 27, two days
after the aborted hearing. The Post-Gazette wrote a next-day story (September
26) about the hearing, as well as another one on Friday, September 27, included
in the present paper. Clearly, the hearing that never was sparked the sustained
interest of the newspaper 40 miles away in Pennsylvania, but for some reason was
not seen as equally important by the hometown newspaper.
     It may be that the Evening Review, as an afternoon paper, did not want to
repeat the lead-sentence information from the Post-Gazette, which would have
come out several hours earlier on any day of the week, thus being able to
potentially 'scoop' its small-town cousin. And it may be that the Review felt it
had even more late-breaking information to report on September 27, because
Martin Sheen's scheduled appearance at the upcoming October protest and rally is
not mentioned in either Post-Gazette article. Even allowing for these
possibilities, it still does not fully explain why the Review would not have
highlighted the fact that several hundred protesters shouted down a hearing
officer in a well-publicized federal hearing that, ultimately, was shut down
before it ever officially got under way.
     The Review's headline-lead structure focus solely on the prospect of
Sheen's return to East Liverpool for a future protest rally, and the next four
paragraphs flesh out the details of this information. Only in paragraph 5 does
the reader learn of the hundreds of opponents shutting down Wednesday night's
federal hearing, and this latter information is contextualized within the
previous information about the upcoming Sheen-led rally: "Next month's rally and
march were announced Wednesday at the Westgate Auditorium, where an estimated
300 WTI opponents began shouting and forced state and federal officials to halt
a hearing into revised a modified permits for the project."
     Another significant difference in the two papers' depictions is the
estimated number of protesters. The Review puts the number at just 300, the
Post-Gazette (in both the September 26 and 27 articles) at 700, more than twice
the former paper's estimate. Van Dijk (1988a) contends that the use of numbers
in news reports is the single most significant discursive feature that anchors
the news in verifiable reality, and which thus lends to the news report an aura
of authoritativeness. Numbers serve a rhetorical or persuasive purpose not only
in their specifying aspects of reality by measurement, but in a more formal
sense, by helping convince readers of the essential real-world quality (or the
"illusion of truth" [van Dijk, 1988a, p. 87) available via the reporting, thus
the trustworthiness of the account. Rhetorically, news discourse must "signal
its credibility and therefore exhibit its truth claims," and it does so through
the conventional use of attributing source materials, presenting factually
verifiable and, especially, numerically based information as a way of
documenting truth claims (van Dijk, 1988a, p. 179.)
     How is it that the two papers can differ so drastically in their estimates
of the number of opponents who chanted and stomped and shut down the event?
Neither paper indicates where their respective estimates came from. One can only
speculate, but the consensus-driven quality of the hometown paper's account of
the hearing - the fact that the 'shouting' event was buried - might mean that
the reporter also would downplay how many opponents turned out to protest. The
Post-Gazette even evaluates the WTI plant's economic value at $12 million more
($140 million) than does the small-town paper, which places the value of the
plant at just $128 million.
     An initial conclusion from the analysis of headline-and-lead structures in
both papers is that the Pittsburgh paper highlighted the conflict of the news
event, while the hometown paper suppressed the significance of the protest, even
though both papers led off with an opposition focus. But in the Pittsburgh
paper's case, the opposition focus was on the event at hand; the small-town
paper emphasized a future protest, and subsumed the more current conflict within
that future occurrence.
     Other Relevant Discourse Features: Wordings, Claims, & Textual Ideology
I will attend only briefly to the interconnected wordings in the texts, to the
rhetorical claims in evidence, and to the ideological problematics that emerge
in each news report. Figures 1, 1a and 2 should, hopefully, go a long way in
demonstrating salient points within each of these areas. I will then conclude by
pointing to the value of the kind of discourse analysis and textual mapping I
have conducted, as well as to some limitations and  implications for future work
along this line.

As an experimental mapping process that has all the markings of an evolving
methodology, the process of connecting arrows to similar words is a pretty messy
affair. Those who favor clear and easy-to-read visual aids will likely be
somewhat put off by the sheer number of criss-crossed lines connected one word
to another, and which do require some patience in following. But the pay-off, I
believe, is that one does at least get a very real sense the lexical density of
even a few paragraphs from a relatively straightforward news story. The real
value is that this lexical mapping allows us to connect the same and similar
words, in a way that demonstrates the redundancy element that is a hallmark of
the mass-mediated text. Semantically, words and their synonyms take up a fair
amount of content-space in the text, and there are ideological implications for
a reading of 'meaning' in the text. At the syntactical level, the mapping
process allows us to make structural connections between sentences and whole
paragraphs, to see how what van Dijk (1988a&b) calls the process of
'discontinuous installment' is at work in a text. Similar units of information
are spread out through the text in installments that do not necessarily follow a
sequential logic.
      The labeling of schema categories [Summary, Verbal Reaction, Context,
Background] merely serves to categorize the narrative functions of distinct
informational units. The designation, 'Topic 2,' in both stories indicates how
after a series of opening paragraphs, a story can veer, sometimes without any
real transition, into a nearly distinct theme. This topic shift, reframing, or
re-footing is similar to what Goffman (1974) has analyzed in face-to-face
conversation. It can be an important element of the news text because it marks
the reporter's transition into what he or she believes is a relevant part of a
larger story, but one that for whatever reasons was not highlighted as the most
important story theme.
     For example, in the Post-Gazette story (Figure 2), Topic 2 has to do with
the nature of "mass public opposition movements, like the one incinerator
opponents are trying to promote," and that these often are successful. The story
then goes into some detail about such movements, quoting a well-known
environmental movement expert. But recall that the main theme or topic of the
story (in the headline, at least) was not about social movements, but about
incinerator 'foes' planning a march and wanting to 'derail' the WTI project.
Recall, as well, that the lead and supporting sentences have nothing to do with
the headline information, nor with the Topic 2 information. Thus, a close
reading of even the first few paragraphs of the story reveals an ideological
contradiction (a problematic) that must be located within the
reporter's/editor's own story-telling logic, even if that logic may be
functioning at an unconscious level.
     One result of identifying an ideological problematic is that it reveals the
reporter's own unique take on reality, typically through description, analytical
and metaphorical language. To 'derail' a project would seem to imply a project
had been sailing along just fine. To be 'chanted down' and being left with 'no
choice but to adjourn a hearing' places responsibility (or irresponsibility)
into the hands of demonstrators. Our hearing officer had 'sat silently on a
middle-school stage,' the picture of calm amid the storm of chaos and protest,
'while opponents chanted and shouted for half an hour.' True, a well-organized
social movement, according to our "expert" quoted in the story, are sometimes
successful in 'derailing' a project - but nowhere in the story does any
protester claim this is his or her mission; yet the headline draws in this word,
supplied by the reporter as a descriptive term, for making a rhetorical claim:
that the 'group' wants to derail the project. In these ways, the ideology of the
powerful (of 'hearing officers' and 'experts') is squared off against the
ideological interests of 'protesters' and 'foes.'
     This same general process is at work in the East Liverpool Evening Review
(figures 1 and 1a) story, only the first several paragraphs of which have been
analyzed here. Though the headline and lead do match, informationally, the
environmental conflict that emerged at the hearing is submerged, and even the
strength of support for the opposition (an estimated 300) is far less than the
estimate of the Post-Gazette. Thus, while the first part of the story is
informationally consistent, the story is nonetheless ideologically problematic.
The ideological kernel in this story is first introduced in paragraph 5, where
the hometown reporter asserts that the 300 WTI opponents 'began shouting and
forced' officials to 'halt a hearing.' Nowhere in this story is any hearing
official quoted as saying that he or she was forced into shutting down the
hearing, though that may indeed be a logical conclusion. But what strikes one,
at first glance, as a denotative, purely descriptive statement is infused, as
well, with a chained metaphor - opponents shouted, and thus 'forced' a 'halt.'
Agency, as in the Post-Gazette account, is clearly on the shoulders of the
protesters. The assumption in the foregoing characterization is that the hearing
is an acceptable process, an orderly, legal, wise action on the part of the
government, and organized by those (state and federal EPA representatives) who
would regulate the incinerator. If one believes the incinerator is not a worthy
or environmentally safe project, or that it has been built with illegal permits,
as opponents repeatedly stressed for years - if this is at least a reasonable
alternative view,  it is nowhere presented in the story.
     Figure 1a does show how the reporter followed his professional duty of
attempting to produce an objective account, by contacting, unsuccessfully, WTI
opponents; and how he reached one opponent who had no comment. Here we see a
significant ideological strategy - from a communicative standpoint, a rhetorical
strategy - at work on both the reporter's and source's side. The reporter tells
readers, in essence, he had tried to get a statement but was unsuccessful,
because the source decided that, strategically, it was not in his best interest
to comment. The reporter's own rhetorical strategy here is to reproduce the
objectivity principle of journalism - that all relevant sides will be included.
The reporter notes not only that one source offered no comment, but that two
other opponents could not be reached. In this way, the news reporter strikes his
own rhetorical stance, that of the objective reporter trying his best, but
failing, to extract information. This is a rhetorical strategy, the subtext of
which is to indicate to readers that objectivity and balance are important
elements of the journalistic text. Thus the reader is given a reason to be
convinced of the essential good faith and interest in accuracy on the part of
the reporter.
     As van Dijk (1988a, p. 179) points out, although news discourse "is
nonpersuasive in principle or intention, it may well have a more persuasive
dimension in a more indirect sense: Even if it does not argue for a position or
opinion, it certainly presupposes them, by definition of its social and
therefore ideological embedding." Here, van Dijk's attention to news discourse
being 'nonpersuasive in principle' points to the traditional journalistic view
that the newspaper should be merely informative, not consciously persuasive or
opinion-rendering. But the second part of the statement, regarding the
'indirect' persuasive dimension of news discourse, which 'presupposes' opinions
because of news discourse's 'social and therefore ideological embedding,' points
exactly to the idea that news reflects the various officially, institutionally
sanctioned, rhetorical positions of powerful social agents. Not least among
these social agents is the news reporter, representing the political-economic
interests of the free-market, advertising-driven mass media institution. Those,
such as environmental advocates, who would challenge these institutions must do
so from outside an institutional logic, which is why these advocates become
metaphorically labeled as protesters and not as hearing officers, officials, or
experts.
     We also encounter, in Topic 2 of the East Liverpool paper's story, a
source-centered claim: if WTI is not permitted to open, no existing industry
will be able to expand. This statement is produced by WTI plant manager Sam
Kasley, and it serves, as well, as a significant topic shift. Space constraints
do not allow further analysis of this newly introduced theme, but it is
significant that the remainder of the article is devoted to virtually one long,
continuing, direct quotation from Kasley. In several long quotations, mediated
only minimally by reporter transitions, Kasley questions, quite effectively, the
environmental logic of those Hollywood types who "can skip town after a march
and fly back to Hollywood" (Evening Review, paragraph, paragraph 11). Kasley is
given an extremely long quotation, in which he asks: Will these Hollywood types
- an obvious reference to Martin Sheen - be willing to stop "sending their
designer clothes to the dry cleaners, stop buying brightly colored outfits, stop
accepting scripts because of the printer's ink on them, stop their servants from
cleaning their homes with cleaning fluids or from using bug sprays to keep
mosquitoes away from their garden parties, stop using batteries in their
limousines and in general reduce their output of hazardous wastes at the
source?"
     A total of 10, mostly lengthy paragraphs are given over completely to
Kasley, virtually all of it in direct quote, a quotation form that typically
takes up relatively little space in most news articles. But perhaps,
rhetorically, the reader is willing to accept this reporting as unbiased because
of the reporter's previous rhetorical strategy, wherein he documented to readers
his noble but failed attempt to get the other side of the story.
     Conclusion
In this paper I have taken a small amount of information, basically the first
several opening paragraphs from two same-day news stories, and expanded upon
them in great detail. The point has been to demonstrate how one might begin to
think through a systematic, structural mapping of the news text in ways that can
clarify both how simple and, at another level, how complex even a relatively
brief news account can be. As well, I have attempted to show how the ideological
interests of both news sources and reporters, respectively representing various
powerful institutions of the general political economy, cohere in the news text
in unique ways. This focus on the reproduction of ideology in texts is an
important part the critical linguistic project. Because journalism is itself a
part of a powerful public-sphere institution (Jensen, 1986) called the mass
media, the journalist naturally defines the news as what happens from within the
operational logic or rhetoric of various other institutions (Fishman, 1980;
Tuchman, 1972, 1978) and most often by presenting an array of officially
sanctioned sources (Ericson et al., 1989; Schlesinger & Tumber, 1994). A
critical discourse-analysis approach to news texts, as I have followed here, is
an attempt to map the discourse of the text and to make critical connections to
the larger social and political world in which and from which the text is
produced.
     There are obvious limits to this approach. It does not lend itself very
easily to analyzing a larger number of items. Even textually mapping just
several paragraphs from a single news story takes significant effort. Therefore,
one is not concerned with questions of generalizability to a larger population,
although the fruits of such analysis are highly valuable for theory building.
The textual mapping approach combined with standard content analysis may be a
useful pairing (see van Dijk, 1983, pp. 25-28). The value of textual mapping as
a close-reading strategy is that one does get a 'depth' reading from a number of
texts, and even in a single text one can begin to unravel ideological
complexities and important narrative chains that allow the analyst to make
verifiable claims about textual meaning. In fact, it may be that textual meaning
- and the ideological implications therein - is not so much (or not only) a
function of semantic analysis but also of how textual structures cohere.  In
other words, tracing the how of meaning, how textual structures emerge in a
text, may be the best way to identify meaning itself. This is not to conclude,
of course, that readers make meaning in the same way. Readers can bring a great
variety of insights into a single analysis of either a text (Liebes & Katz,
1990) but it is also true that textual structures inherently limit or at least
anchor any individual toward certain fundamental meanings that are 'in' the text
(Condit, 1989).
     In discourse analysis, there is no clear-cut distinction between what
counts as theory and what as method. As outlined by van Dijk, for example,
discourse analysis is both a theory of how ideology is realized through language
and a method for analyzing it. One is always engage in theoretical exploration
and methodological exposition. Van Dijk's statement that "a serious and critical
analysis of the ideological dimension of news is impossible without this kind of
explication of the links that bind news structures with the social cognitions of
journalists as members in ideological institutions such as the media," gets at
the heart of my own theoretical logic informing the textual mapping schema
demonstrated in the present paper.
      My own mapping system, which is most influenced or borrows from
functional-linguistics (Martin, 1992; Halliday, 1985), from news discourse
analysis (van Dijk, 1988a&b, 1985, 1991, 1993) and from critical linguistics
(Fowler, 1991; Fowler et al., 1979), is an amalgamated approach to linguistic
analysis. It is my own still-experimental and somewhat messy attempt to think
through textual mapping in a way that will in the future provide somewhat more
parsimonious results.
     The hope is that with further explication, this brand of news discourse
analysis might become of greater use in general mass media textual studies. As
outlined here, the method can be theorized as a rough-and-ready attempt to tap
into the ideological meaning-structure of the text.





















[1]  I use or adopt the following schema categories from van Dijk. Topic is a
combination of the headline and lead sentence(s), and which functions to provide
an overall summary of what the newspaper reporter finds most relevant about a
given news event or issue; Context provides relevant immediate information
within which a current news issue or event can be understood; Background
provides more distant but still fairly recent past information, which gives a
more historical context to the immediate news event or issue; History is
information from the more-distant past but relevant to understanding a current
event; Verbal Reaction is the direct and indirect quoting of news sources;
Consequences is reporter-centered analysis that provides a causal explanation,
or informed judgment about the event at hand; Comment is the express opinion or
overt evaluation - often through just a word or two, other times through one or
more sentences - offered, consciously or not, by a reporter. In the present
article, only the more frequently occurring schema categories of Topic, Context,
Background, and Verbal Reaction are explored. For explication of the schema
categories, see van Dijk (1885; 1988a&b; 1991a&b; 1993.) My own work (Harry,
1999) explores these categories in much greater detail, as well, and applies
them to a wide range of comparative news texts.
[2]  In Harry (1999), I analyze the structural-pluralism hypothesis of Tichenor
et al., as most fully documented in their Community Conflict & the Press (1980),
in the interest of assessing its goodness-of-fit with respect to the Pittsburgh
and East Liverpool newspaper cases.  I do not address that level of analysis in
the present paper, which is more concerned with comparative linguistic
structures in the news reports.
[3]  As a historical note, the WTI incinerator went into limited commercial
operation on April 12, 1993, about 18 months after the Sept. 27, 1991, news
event analyzed in the present paper. Well-organized protests and sustained
coverage occurred in both newspapers until the plant opened, and then dropped
off dramatically afterward. Since then, only minimal coverage of WTI has been
produced by either newspaper.